Evaluate the incidence and nature of differences in AMNOG assessment outcomes between IQWiG and G-BA, identify reasons and assess potential drivers of discrepancy between G-BA and IQWiG.

Collection and analyse of outcomes of 283 finalized G-BA non-orphan drug assessments (by January 2019) and identified those where G-BA and IQWiG outcomes differed, either by G-BA granting more favourable benefit assessment outcome than IQWiG or vice versa. Reasons for the decision making were analysed and categorized in both cases, and what triggered the change in outcome. We also investigated a correlation of such discrepancies with certain therapeutic areas.

By January 2019, 283 assessments were conducted/finalized by G-BA. Out of those, 50 outcome assessments varied between IQWiG and G-BA in at least one subpopulation. In 43 cases, G-BA opinion differed from IQWiG and an additional benefit rating was granted in at least one subpopulation. In 7 cases, G-BA did not confirm IQWiG rating and did not grant additional benefit. IQWiG oftentimes based the denial of additional benefit on reasons such as manufacturer-biased evidence, data inadequacy, discrepancies between label and trial, lack of evidence or choice of an inappropriate comparator. In 16 cases, G-BA overruled IQWiG decision based on input from the commenting procedure. Correlation between types of reasons for IQWIG and therapeutic areas was observed. Use of an inappropriate comparator was the sole reason overruled by G-BA in metabolic disorders. Oncology and infectology presented a wide range of reasons why IQWIG did not recognize a benefit, but was overruled by G-BA.

There is a number of cases where G-BA overruled IQWiG decision based on updated data or input during the commenting procedure. This implies that manufacturers, in case of an unfavourable IQWiG decision, need to be prepared to deliver the right data subsequently and provide input from clinical experts in the commenting procedure.